Thursday, May 26, 2011

Sex Researcher Calls for "An Evidence-Informed Process"

Brian Mustanski, Associate Professor, Department of Medical Social Sciences, Feinberg School of Medicine, Northwestern University, calls for "moving the IRB process of risk/benefit assessment from being entirely subjective to being evidence-based."

[Brian Mustanski, "Ethical and Regulatory Issues with Conducting Sexuality Research with LGBT Adolescents: A Call to Action for a Scientifically Informed Approach," Archives of Sexual Behavior, published online 29 April 2011.]

Mustanski is particularly conerned with research on lesbian, gay, bisexual, and transgender (LGBT) youth. Such youth, especially men who have sex with men (MSM), are at elevated risk of illness, especially HIV infection. But researchers have done relatively little to study interventions targeted at adolescent MSM. More generally, scholars' failure to study LGBT adolescents weakens "our basic understanding of diversity in the development of core aspects of sexuality, like sexual attractions and orientation, gender identity and expression, and romantic relationships."

Why this lack of research? Mustanski (a former IRB member himself) believes that IRBs are a big part of the problem. He cites studies showing that psychologists have modified or abandoned plans for fear of IRB rejection.

As he puts it,

It can be extremely frustrating to agonize over perfecting every detail of a protocol to only then have it repeatedly questioned by anonymous individuals who may not have the same subject matter expertise. The standard IRB approach of written correspondence with long delays can be exasperating and is not designed to facilitate a collaborative relationship between IRBs and investigators. Investigators may be disturbed by requirements to make protocol changes that they perceive as decreasing the quality of the science without apparent reduction in risks or increasing protections. Communication can sometimes be unclear and, if a board has limited experience with the LGBT community, questions may seem uninformed, insensitive, or even homophobic. Community partnerships may be strained by the need for investigators to comply with IRB mandates over the recommendations of community members with expertise in serving the target population.

Mustanski notes his own involvement on a project that was funded for two years; IRB approval ate ten of those twenty-four months. Among the problems:

  • The IRB applauded portions of the application in one round, then found those same portions unsatisfactory in a following round.
  • The IRB labeled LGBT people as a vulnerable population, despite the fact that federal regulations do not list them as such. Mustanski notes that the regulatory categories are based on groups' inability to make decisions based on cognitive disability or a lack of power. "I know of no evidence that demonstrates such decisional impairment" with LGBT people, he writes, "and I believe many LGBT individuals would be insulted to have it implied otherwise."
  • The IRB required the consent process to include the warning that some questions "could make you feel uneasy or embarrassed" and that counseling services would be available. But 89 percent of those interviewed found the experience no more uncomfortable than "a typical visit to your physician, doctor, psychologist, or counselor."
One could see why researchers would be tempted to achieve less scrutiny by limiting their surveys to respondents over age 18, or agreeing to parental consent for studies of younger people. But Mustanski warns that either approach would shortchange communities. Younger adolescents face challenges that their 18-year-old counterparts will remember less clearly, or--because of the changing culture--that 18-year-olds did not experience in the same way. And a "parental consent requirement will tend to underrepresent many of the youth who are most in need of the benefits of research, such as the development and validation of targeted health programs."

To improve the chances of future research making it through the IRB with less difficulty, Mustanski offers a list of recommendations to investigators. Mostly he thinks that if investigators become expert in the empirical data about risks and benefits of their research as well as relevant federal and local laws and regulations, and offer to share that information in in-person meetings with IRB staff and the board itself, everything will be fine.

I do hope so, though I fear that these recommendations may overlook the power dynamic involved. I am reminded of the 2007 University of California Academic Council Report on Institutional Review Boards at UC:
The challenge of training faculty IRB members is exacerbated by the extreme time commitment of serving on the IRB. There is little time available to faculty to be trained on subject protection beyond the time committed to protocol review. Some campuses include training in the IRB meetings, devoting 5 to 15 minutes of meeting time to developments in subject protection. However, IRB staff report that when training is on the agenda of the IRB meeting, faculty members often skip that part of the meeting because they are so busy. Rarely is there funding to train faculty IRB members.
Mustanski offers no mechanism by which IRBs would be required to deploy or even read the information he wants investigators to compile. For example, he could have endorsed the 1999 proposal of the Working Group of the Human Subjects Research Subcommittee of the National Science and Technology Council that:
In determining whether there might be a reasonable risk or damage related to divulging the sensitive information, etc., it is not enough that there be merely some hypothetical possible risk that can be construed. Rather, the risks resulting from disclosure must be readily appreciable and significant.
Still, even a cynic can join a call for emprical data, which could be useful to researchers even if IRBs fail to do their homework. So my favorite recommendation is number three. Researchers should
help advance knowledge about the risks and benefits of participating in research by asking participants how they felt about their participation in your study. Publish the results so as to build a corpus of knowledge that will allow for evidence-based determinations about risks and benefits. As scientists, we are in a unique position to bring to bear our expertise in understanding how people respond in various situations and we should harness this expertise to help inform the IRB review process. Whenever possible, provide your IRB the empirical data about risks and benefits of sexuality research as they may be unfamiliar with this area.
Whatever one's feelings about IRBs, empirical research on the risks and benefits of different types of research is good for everyone. In some cases, this will take the form of projects designed specifically to measure the risks and benefits of research; many articles in the Journal of Empirical Research on Human Research Ethics take this form. But Mustanski's experience suggests that researchers who do not set out primarily to advance knowledge of research ethics nonetheless can accumulate valuable data as a byproduct of their main endeavor, e.g., by asking a question or two about how participants felt about participating.

Gathering and publishing such information (even informally, as on a blog), could indeed contribute to Mustanski's goal of improving IRB review "by transforming it into an evidence-informed process." And even if it doesn't, researchers would be better able to make their own ethical decisions.

Mustanski ends his article with sample language that could be included in an IRB application for a study of under-18 youth where parental consent is not being sought. This follows Rena Lederman's call for IRB boilerplate. Of course, when someone actually tried Lederman's boilerplate for fieldwork, the IRB responded by insisting on its own standard procedures. Maybe it's the thought that counts.


Anonymous said...

Our institution has been conducting this type of research for a couple of decades. Getting this research through the IRB isn't a huge deal. Yes, there are generally more concerns about these type of studies, given the nature of the information being collected, but I don't think this adds much to the review time.

If we are going to talk about evidence, where's the evidence that IRBs are a major barrier to doing this type research? CDC, NIH and other agencies have spent billions of dollars on behavioral and social science research associated with the prevention of STDs, HIV, etc. This research isn't being done or is suffering serious delays because of IRBs? Where's the evidence?

Zachary M. Schrag said...

In addition to his own experiences, Professor Mustanski offers two citations for his claim that research on LGBT youth is hampered by IRB review.

* Fendrich, M. (2009, August). Empirical research on risks from studies of sexuality and other sensitive topics: What we don’t know may be hurting us. Paper presented at the meeting of the International Academy of Sex Research, San Juan, Puerto Rico.

* Miller, R. L., Forte, D., Wilson, B. D., & Greene, G. J. (2006). Protecting sexual minority youth from research risks: Conflicting perspectives. American Journal of Community Psychology, 37, 341–348.

I don't think he would dispute that "CDC, NIH and other agencies have spent billions of dollars on behavioral and social science research associated with the prevention of STDs, HIV, etc." His complaint is that much of this funding has been directed at adult populations that are at lower risk than the adolescents he wishes to help.

Anonymous said...

Look up the numbers and you'll find a lot of money is spent on adolescent public health, including HIV prevention in high risk groups such as YMSM.

Zachary M. Schrag said...

Thanks for this comment and for the link to the CDC's page on "Lesbian, Gay, Bisexual and Transgender Health."

I did not see on that page any data indicating the levels of federal support for surveys of under-18 adolescents conducted without parental consent.

I did see a May 2009 fact sheet on HIV/AIDS and Young Men Who Have Sex with Men. That fact sheet, in turn, features a 14-item bibliography.

The items that describe surveys of under-18 adolescents include:

* MacKellar DA, Valleroy L, Secura G, et al. Unrecognized HIV infection, risk behaviors, and perceptions of risk among young men who have sex with men: opportunities for advancing HIV prevention in the third decade of HIV/AIDS. J AIDS 2005; 38:603–614.

This is based on data from the CDC's Young Men's Survey (YMS) which surveyed men under the age of 23 only in 1994-1998.

* Garafolo R, Wolf RC, Kessel S, Palfrey J, DuRant RH. The association between health risk behaviors and sexual orientation among a school-based sample of adolescents. Pediatrics 1998;101:895–902.

This one was based on the the 1995 Massachusetts Youth Risk Behavior Survey (YRBS). Students needed parental consent to participate (though it is true that "fewer than 10 students were denied parental permission.")

* Goodenow C, Netherland J, Szalacha L. AIDS-related risk among adolescent males who have sex with males, females, or both: evidence from a statewide survey. American Journal of Public Health 2002;92:203–10.

"Study participants were sexually experienced males from the sample of high school students who completed the 1995, 1997, or 1999 Massachusetts Youth Risk Behavior Survey (MYRBS)." Parental permission was required to participate.

* Robin L, Brener ND, Donahue SF, Hack T, Hale K, Goodenow C. Associations between health risk behaviors and opposite-, same-, and both- sex sexual partners in representative samples of Vermont and Massachusetts high school students. Archives of Pediatric and Adolescent Medicine 2002;156:349–55.

Based on 1995 and 1997 data from the Vermont and Massachusetts Youth Risk Behavior Surveys. The article does not address the issue of parental consent, but the previous two citations indicate that parental consent was needed to participate in at least the Massachusetts version.

It seems, then, that the CDC is basing its advice to educators and school administrators on a small set of data from surveys conducted in the 1990s, most of which required parental consent to participate. Since IRBs became far more strict after the federal Office for Protection from Research Risks suspended funding at major institutions starting in 1998, it is plausible that new IRB rules are responsible for the absence of more recent data.